Controversial Study Investigates Therapeutic Benefit of Placebo

September 1, 1998
Volume 15, Issue 9

How much of the beneficial effects of anti-depressant medications can be ascribed to the placebo effect? Irving Kirsch, Ph.D., and Guy Sapirstein, Ph.D., addressed this important question in a recent study that appeared in the first volume of the American Psychological Association's online journal, Prevention and Treatment (June 26,1998). Although their methodology and conclusions have met with some controversy, it would be imprudent to invalidate the study and its hypothesis.

How much of the beneficial effects of anti-depressant medications can be ascribed to the placebo effect? Irving Kirsch, Ph.D., and Guy Sapirstein, Ph.D., addressed this important question in a recent study that appeared in the first volume of the American Psychological Association's online journal, Prevention and Treatment (June 26,1998). Although their methodology and conclusions have met with some controversy, it would be imprudent to invalidate the study and its hypothesis.

Kirsch and Sapirstein calculated the mean effect sizes for 2,318 patients who were randomly assigned to antidepressant medication or placebo groups in 19 double-blind clinical trials. Of the total 2,318 participants, 1,460 received medication and 858 received placebo. The results of the studies were combined through the use of a meta-analysis. To delineate the components of the placebo effect, the authors utilized psychotherapy outcome studies.

The authors divided the 16 medications used in the various studies into four general categories: tricyclics and tetracyclics, selective serotonin reuptake inhibitors, other antidepressants and other medications not considered antidepressants (amylobarbitone, lithium [Lithobid, Eskalith], liothyronine [Cytomel, Triostat], and adinazolam).

From their data, the authors concluded that "75% of the response to the medications examined in these studies was a placebo response and, at most, 25% might be a true drug effect...[that] means that for a typical patient, 75% of the benefit obtained from the active drug would have also been obtained from an inactive placebo."

Kirsch and Sapirstein attempted to explain the overwhelming efficacy of nonantide-pressant medications. Perhaps, they suggested, these medications are affecting depression indirectly by improving sleep, lowering anxiety and the like. Their effects should have been less than that of true antidepressants, the authors wrote; instead the effect observed was "at least as great as that to conventional antidepressants."

Another possible answer is that antidepressant medication might function as active placebo, in which the side effects amplify the placebo effect by convincing patients that they are receiving a potent drug.

Natural history effects also may play a role in the magnitude of placebo effects observed. The authors note that "the promise of future treatment is sufficient to trigger a placebo response." Likewise, the assignment to a control group can cause feelings of hopelessness and, accordingly, increase feelings of depression.

The authors recommend four-arm studies to ascertain definitive effects of active placebos, inactive placebos, active medication and natural history-HAD